Virtual humans are computer-generated characters designed to look and behave like real people. Studies have shown that virtual humans can mimic many of the social effects that one finds in human-human interactions such as creating rapport, and people respond to virtual humans in ways that are similar to how they respond to real people. We believe that virtual humans represent a new metaphor for interacting with computers, one in which working with a computer becomes much like interacting with a person and this can bring social elements to the interaction that are not easily supported with conventional interfaces. We present two systems that embody these ideas. The first, the Twins are virtual docents in the Museum of Science, Boston, designed to engage visitors and raise their awareness and knowledge of science. The second SimCoach, uses an empathetic virtual human to provide veterans and their families with information about PTSD and depression.

Intelligent game-based learning environments integrate commercial game technologies with AI methods from intelligent tutoring systems and intelligent narrative technologies. This article introduces the CRYSTAL ISLAND intelligent game-based learning environment, which has been under development in the authors’ laboratory for the past seven years. After presenting CRYSTAL ISLAND, the principal technical problems of intelligent game-based learning environments are discussed: narrative-centered tutorial planning, student affect recognition, student knowledge modeling, and student goal recognition. Solutions to these problems are illustrated with research conducted with the CRYSTAL ISLAND learning environment.

Articulating thought in computer-based media is a powerful means for humans to develop their understanding of phenomena. We have created DynaLearn, an Intelligent Learning Environment that allows learners to acquire conceptual knowledge by constructing and simulating qualitative models of how systems behave. DynaLearn uses diagrammatic representations for learners to express their ideas. The environment is equipped with semantic technology components capable of generating knowledge-based feedback, and virtual characters enhancing the interaction with learners. Teachers have created course material, and successful evaluation studies have been performed. This article presents an overview of the DynaLearn system.

We study the power of different types of adaptive (nonoblivious) adversaries in the setting of prediction with expert advice, under both full-information and bandit feedback. We measure the player's performance using a new notion of regret, also known as policy regret, which better captures the adversary's adaptiveness to the player's behavior. In a setting where losses are allowed to drift, we characterize ---in a nearly complete manner--- the power of adaptive adversaries with bounded memories and switching costs. In particular, we show that with switching costs, the attainable rate with bandit feedback is $T^{2/3}$. Interestingly, this rate is significantly worse than the $\sqrt{T}$ rate attainable with switching costs in the full-information case. Via a novel reduction from experts to bandits, we also show that a bounded memory adversary can force $T^{2/3}$ regret even in the full information case, proving that switching costs are easier to control than bounded memory adversaries. Our lower bounds rely on a new stochastic adversary strategy that generates loss processes with strong dependencies.

We study the power of different types of adaptive (nonoblivious) adversaries in the setting of prediction with expert advice, under both full-information and bandit feedback. We measure the player's performance using a new notion of regret, also known as policy regret, which better captures the adversary's adaptiveness to the player's behavior. In a setting where losses are allowed to drift, we characterize ---in a nearly complete manner--- the power of adaptive adversaries with bounded memories and switching costs. In particular, we show that with switching costs, the attainable rate with bandit feedback is $T^{2/3}$. Interestingly, this rate is significantly worse than the $\sqrt{T}$ rate attainable with switching costs in the full-information case. Via a novel reduction from experts to bandits, we also show that a bounded memory adversary can force $T^{2/3}$ regret even in the full information case, proving that switching costs are easier to control than bounded memory adversaries. Our lower bounds rely on a new stochastic adversary strategy that generates loss processes with strong dependencies.

The increased availability of data in recent years led several authors to ask whether it is possible to use data as a {\em computational} resource. That is, if more data is available, beyond the sample complexity limit, is it possible to use the extra examples to speed up the computation time required to perform the learning task? We give the first positive answer to this question for a {\em natural supervised learning problem} --- we consider agnostic PAC learning of halfspaces over $3$-sparse vectors in $\{-1,1,0\}^n$. This class is inefficiently learnable using $O\left(n/\epsilon^2\right)$ examples. Our main contribution is a novel, non-cryptographic, methodology for establishing computational-statistical gaps, which allows us to show that, under a widely believed assumption that refuting random $\mathrm{3CNF}$ formulas is hard, efficiently learning this class using $O\left(n/\epsilon^2\right)$ examples is impossible. We further show that under stronger hardness assumptions, even $O\left(n^{1.499}/\epsilon^2\right)$ examples do not suffice. On the other hand, we show a new algorithm that learns this class efficiently using $\tilde{\Omega}\left(n^2/\epsilon^2\right)$ examples. This formally establishes the tradeoff between sample and computational complexity for a natural supervised learning problem.

We study the problem of online learning in finite episodic Markov decision processes (MDPs)where the loss function is allowed to change between episodes. The natural performance measure in this learning problem is the regret defined as the difference between the total loss of the best stationary policy and the total loss suffered by the learner. We assume that the learner is given access to a finite action space A and the state space X has a layered structure with L layers, so that state transitions are only possible between consecutive layers. We describe a variant of the recently proposed Relative Entropy Policy Search algorithm and show that its regret after T episodes is 2 L X A T log( X A /L) in the bandit setting and 2L T log( X A /L) in the full information setting, given that the learner has perfect knowledge of the transition probabilities of the underlying MDP. These guarantees largely improve previously known results under much milder assumptions andcannot be significantly improved under general assumptions.

We consider the problem of learning good trajectories for manipulation tasks. This is challenging because the criterion defining a good trajectory varies with users, tasks and environments. In this paper, we propose a co-active online learning framework for teaching robots the preferences of its users for object manipulation tasks. The key novelty of our approach lies in the type of feedback expected from the user: the human user does not need to demonstrate optimal trajectories as training data, but merely needs to iteratively provide trajectories that slightly improve over the trajectory currently proposed by the system. We argue that this co-active preference feedback can be more easily elicited from the user than demonstrations of optimal trajectories, which are often challenging and non-intuitive to provide on high degrees of freedom manipulators. Nevertheless, theoretical regret bounds of our algorithm match the asymptotic rates of optimal trajectory algorithms. We also formulate a score function to capture the contextual information and demonstrate the generalizability of our algorithm on a variety of household tasks, for whom, the preferences were not only influenced by the object being manipulated but also by the surrounding environment.

This paper addresses the problem of online learning in a dynamic setting. We consider a social network in which each individual observes a private signal about the underlying state of the world and communicates with her neighbors at each time period. Unlike many existing approaches, the underlying state is dynamic, and evolves according to a geometric random walk. We view the scenario as an optimization problem where agents aim to learn the true state while suffering the smallest possible loss. Based on the decomposition of the global loss function, we introduce two update mechanisms, each of which generates an estimate of the true state. We establish a tight bound on the rate of change of the underlying state, under which individuals can track the parameter with a bounded variance. Then, we characterize explicit expressions for the steady state mean-square deviation(MSD) of the estimates from the truth, per individual. We observe that only one of the estimators recovers the optimal MSD, which underscores the impact of the objective function decomposition on the learning quality. Finally, we provide an upper bound on the regret of the proposed methods, measured as an average of errors in estimating the parameter in a finite time.

All the existing multi-task local learning methods are defined on homogeneous neighborhood which consists of all data points from only one task. In this paper, different from existing methods, we propose local learning methods for multi-task classification and regression problems based on heterogeneous neighborhood which is defined on data points from all tasks. Specifically, we extend the k-nearest-neighbor classifier by formulating the decision function for each data point as a weighted voting among the neighbors from all tasks where the weights are task-specific. By defining a regularizer to enforce the task-specific weight matrix to approach a symmetric one, a regularized objective function is proposed and an efficient coordinate descent method is developed to solve it. For regression problems, we extend the kernel regression to multi-task setting in a similar way to the classification case. Experiments on some toy data and real-world datasets demonstrate the effectiveness of our proposed methods.